Diesel Engine Instantaneous Oil Consumption Measurements Using The Sulfur Dioxide Tracer Technique

Diesel Engine Instantaneous Oil Consumption Measurements Using the Sulfur
Dioxide Tracer Technique
by
Douglas M. Schofield
B. S., Marine Engineering and Naval Architecture; United States Coast Guard Academy,
1991
Submitted to the Departments of Ocean Engineering and Mechanical Engineering in

Partial Fulfillmentof the Requirements for the Degrees of
MASTER OF SCIENCE IN NAVAL ARCHITECTURE AND MARINE

ENGINEERING
and
MASTER OF SCIENCE IN MECHANICAL ENGINEERING
at the
Massachusetts Institute of Technology; May 1995
© 1995 Douglas M. Schofield. All rights reserved.

The author hereby grants to MIT permission to reproduce and to distribute publicly paper
and electronic copies of this doqument in whole or in part.
Signature of Author __Y__ I_ __ · )__I___

-,Dep#artment of Ocean Engintering May 1995
Certified by
Dr. Alan J. Brown
Profe'ssor, Department of Ocean Engineering
Thesis Reader
Certified byJ
Dr2Aictor W. Wong
Lecturer, Department of Mechanical Engineering
Thsis Advisor
Accepted by v * ~~~~- -g
eg7
-'--- A. Douglas Carmichae, Chairman
Departmental Committee on Graduate Studies
Departnme.Qoef-Qqn Engineering
Accepted by
MASSACHUS;TTS INSTITUTE
A. A. Sonin, Chairman
OF TECHNOLOGY Departmental Committee on Graduate Studies
Department of Mechanical Engineering
JUL 2 8 1995
LIBRARIES
1
" ARCHI\VES
(This page intentionally left blank)
2
Diesel Engine Instantaneous Oil ConsumptionMeasurements Using the Sulfur Dioxide Tracer
Technique
by
Submitted to the Department of Ocean Engineering and the Department of Mechanical Engineering in
Partial Fulfillment of the Requirements for the Degrees of Master of Science in Naval Architecture and
Marine Engineering and Master of Science in Mechanical Engineering.
ABSTRACT
A study of the oil consumption characteristics was completed using a single cylinder direct
injection diesel engine. The objectives included establishing the testing procedures for a real-time sulfur
dioxide oil consumption system and studying the ring-pack derived oil consumption including testing the
repeatability and variability through load changes, speed changes, and changes in oil control ring tension.
Also, the use of a commercially available oil was investigated to determine its effectiveness in a sulfur
dioxide testing apparatus.
The testing included three test matrices all completed at steady-state engine operating conditions.
The first had two loading conditions through three speeds while continuously monitoring the ring-pack oil
consumption. This matrix had 24 tests while using a standard oil control ring. The second had the same
two loading conditions through two speeds while continuously monitoring only the ring-pack oil
consumption. This matrix had 22 tests using an oil control ring with 45% less tension than the standard
configuration. The third matrix had two loading conditions through two speeds with 12 tests. This
matrix measured the entire engine oil consumption through the same apparatus, but the engine oil was a
commercially available 10W-30 grade oil.
Several conclusions were drawn from the data analysis. The first matrix showed that the oil
consumption data was repeatable from test to test for the high and low speed, but the mid-range speed
showed oil consumption variability with different oil consumption plateaus lasting between 15 and 30
minutes with changes in oil consumption among these plateaus as much i00%. These plateaus may be
caused by the axial and rotational dynamics of the piston rings. The second matrix results showed larger
variability with instantaneous spiking of the oil consumption occurring during the continously monitored
real-time oil consumption. These spikes were the rapid rising and falling of oil consumption with
increases as much as 500% and often periodic in behavior with periods from 2 to 20 minutes. The
instantaneous spikes may be caused by ring-pack reverse blow-by gases, transporting oil into the cylinder
combustion region. Together, the analyses showed that by decreasing the oil control ring tension by 45%,
the mostly ring-pack derived oil consumption did not change significantly at low speed but increased by
63% at the maximum speed. Also, by decreasing the oil control ring tension, more variability in the
mostly ring-pack derived oil consumption values were seen. The third matrix showed that this sulfur
dioxide system was sensitive enough to determine the oil consumption values at the 1200 and 2400 rpm
using a commercially available 10W-30 oil.
Thesis Supervisor: Victor W. Wong

Title: Lecturer, Department of Mechanical Engineering
Thesis Reader: Alan J. Brown
Title: Professor of Naval Construction and Engineering
3
4
ACKNOWLEDGMENTS
The success of this project was by no means an individual effort. I was given
support from many people, and I would like to sincerely thank them all.
Eric Ford has contributed a lot to this project. His patience and attitude made
working together easy and fun. His organization and devotion to the task at hand
influenced all people involved in the process.
My advisors, Captain Brown and Dr. Wong, both provided the efficient means in
which to complete the current study. Other faculty members at MIT gave crucial
comments and guidance in the successful completion of this thesis. Their guidance and
support were extremely beneficial; I thank you all.
Dr. Alan Bentz of the USCG R&D Center was a critical asset to the project. His
financial support and chemistry technical advice were invaluable. His guidance has not
only made my research go smoothly, but has also supported many other Coast Guard and
Naval Officers here at MIT.
Richard Flaherty at Cummins Engine Company gave many hours of his time in
setting up the oil consumption system. His patience and determination of the project were
most valuable. Throughout the entire extended dyno repair period, he continued to push
forward, making the project meet its deadline. He was truly a mentor, whom I would feel
honored to work for in the future.
Many others contributed efforts towards this project. David Fiedler of Dana
Perfect Circle, Philip Burnett of Shell Oil, and Norbert Abraham of Cummins Engine
Company, all contributed critical pieces to the overall project.
The members of the Sloan Lab Staff also deserve recognition. Don Fitzgerald
gave excellent technical and ordering support. Nancy Cook gave important administrative
assistance. Brian Corkum gave excellent "hands-on" guidance and skill. His devotion to
always teaching new students the tools and methods of the lab is irreplaceable. Without
his support, the project never would have left the office.
5
Office mates and friends always make projects go a little easier. R.B., Erik,
Brigitte, Dan, Rob, Mark, Lee, Tom, Jim, Mary-Ann, Goro, Tian, Denis all made my time
here interesting and fun. I thank you all for your support and friendship.
Finally, my wife deserves the most credit for my work. Her support through high
phone bills kept me on track for the task at hand. Without her support, the project would
not have been completed.
This work has been supported by the Consortium Of Lubrication in Internal

Combustion Engines, whose members include Ford Motor Company, Shell Oil Company,
Dana Corporation, Peugeot, Renault, and Pennzoil Products, and also by the United
States Coast Guard and CumminsEngine Company.
May, 1995
6
TABLE OF CONTENTS
ABSTRACT 3
ACKNOWLEDGMENTS 5
TABLE OF CONTENTS 7
LIST OF FIGURES 9
LIST OF TABLES 11
ABBREVIATIONS 13
Chapter 1 MOTIVATION 15
1.1 Introduction 15
1.2 Previous Work 15
1.3 Oil Consumption Sources 16
1.4 Oil Consumption Measurement Techniques 17
1.5 Objectives 18
Chapter 2 EXPERIMENTAL APPARATUS 19

2.1 Engine 19
2.2 Oil Consumption Apparatus 22
Chapter 3 EXPERIMENTATION 25
3.1 General Overview Of Testing 25
3.2 Testing Procedure For The Standard Oil Control Ring Tension 26
3.21 Test Matrix A 26
3.22 Procedure 26
3.23 Matrix A Engine Specifics, Controls, and Variables 28
3.3 Testing Procedure For The Less Tensioned Oil Control Ring 30
3.31 Test Matrix B 30
3.32 Procedure 31
3.33 Matrix B Engine Specifics, Controls, and Variables 32
3.4 Testing Procedure For The CommerciallyAvailable Oil 32
3.41 Test Matrix C 32
3.42 Procedure 33
7
3.43 Matrix C Engine Specifics, Controls, and Variables 33
Chapter 4 THEORY 34
4.1 General 34
4.2 Matrix Application 37
Chapter 5 RESULTS AND ANALYSIS 38
5.1 Results For The Standard Oil Control Ring 38
5.11 Matrix A Steady-State Oil Consumption 38
5.12 Variability of Matrix A During Steady-State Operation 39
5.2 Results For The Less Tensioned Oil Control Ring 40
5.21 Matrix B Steady-State Oil Consumption 40
5.22 Variability Of Matrix B During Steady-State Operation 41
5.3 Results For The Commercially Available Oil 42
5.31 Matrix C Steady-State Oil Consumption 42
5.32 Variability of Matrix C During Steady-State Operation 43
5.4 Comparison Of Matrices A, B, C 44
5.41 Ring-Pack OC Evaluation 44
5.42 Effects Of The Less Tensioned Oil Control Ring 44
5.43 Comparison Of Commercially Available Oil to High Sulfur Oil 45
Chapter 6 CONCLUSIONS 46
Chapter 7 RECOMMENDATIONS 48
REFERENCES 49
Appendix A TESTING RECORDING SHEETS 51
Appendix B MATRIX A RESULTS 53
Appendix C MATRIX A REAL-TIME OC RESULTS 56
Appendix D MATRIX B RESULTS 63
Appendix E MATRIX B REAL-TIME OC RESULTS 68
Appendix F MATRIX C RESULTS 73
Appendix G MATRIX C REAL-TIME OC RESULTS 76
Appendix H COMPARISON OF MATRICES 79
8
LIST OF FIGURES
Figure 2-1 Exhaust System 21

Figure 2-2 Valve Oil System 22
Figure 2-3 OC Sampling System 23
Figure A-i Calibration Data 51
Figure A-2 Testing Data Sheet 52

Figure B- Matrix A Data 53
Figure B-2 Matrix A Steady-State OC Results 54

Figure B-3 Matrix A Steady-State SOC Results 55
Figure C- Matrix A Condition A: 1200 Rpm Low Load 56
Figure C-2 Matrix A Condition B: 1200 Rpm High Load 57
Figure C-3 Matrix A Condition C: 2400 Rpm Low Load 58
Figure C-4 Matrix A Condition E: 3300 Rpm Low Load 59
Figure C-5 Matrix A Condition F: 3300 Rpm High Load 60
Figure C-6 Chart Recording Of Condition C3 61
Figure C-7 Chart Recording Of Condition C2 62

Figure D-1 Matrix B Data 63
Figure D-2 Matrix B Steady-State OC Results Condition I, J, L 64
Figure D-3 Matrix B Steady-State OC Results Conditions T, U 65
Figure D-4 Matrix B Steady-State SOC Results Conditions I, J, L 66
Figure D-5 Matrix B Steady-State SOC Results Conditions T, U 67
Figure E- Matrix B Condition I: 1200 Rpm Low Load 68
Figure E-2 Matrix B Condition J: 1200 Rpm High Load 69

Figure E-3 Matrix B Condition L: 3300 Rpm High Load 70
Figure E-4 Matrix B Condition T: 1200Rpm High Load 71
Figure E-5 Matrix B Condition U: 3300 Rpm High Lcd 72
Figure F- 1 Matrix C Data 73
Figure F-2 Matrix C Steady-State OC Results 74

Figure F-3 Matrix C Steady-State SOC Results 75
9
Figare G-l Matrix C Condition REGA: 1200 Rpm Low Load 76
Figure G-2 Matrix C Condition REGB: 1200 Rpm High Load 77
Figure G-3 Matrix C Condition REGF: 3300 Rpm High Load 78
Figure H-1 Total Engine SOC to Ring-Pack Comparison 79
Figure H-2 Oil Control Ping Tension Comparison 80
Figure H-3 High Sulfur and Commercially Available Oil Comparison 81
10
LIST OF TABLES
Table 2-1 Engine Characteristics 19
Table 2-2 Engine Thermocouples 20

Table 3-1 Overview of Testing 25
Table 3-2 Test Matrix A 26
Table 3-3 Diesel Fuel Characteristics 29
Table 3-4 Oil Characteristics 29
Table 3-5 Piston Ring-Pack Characteristics 30

Table 3-6 Matrix A Variable Specifics 30
Table 3-7 Test Matrix B 31
Table 3-8 Modified Oil Control Ring Characteristics 32

Table 3-9 Test Matrix C 33
Table 3-10 Commercially Available Oil Characteristics 33
Table 5-1 Summary of Matrix A Steady-State SOC Results 38
Table 5-2 Summary of Matrix B Steady-State SOC Results 40
Table 5-3 Summary of Matrix C Steady-State SOC Results 43
Table 5-4 Valve Oil Contribution 44

Table 5-5 Matrix A to B SOC Comparison 44
Table 5-6 Matrix C to High Sulfur Oil SCC Comparison 45
11
12
ABBREVIATIONS
SYMBOL DEFINITION UNITS
a # of carbon atoms per fuel molecule

AiRvolt Meter voltage output due to sulfUr in ambient air volt
b # of hydrogen atoms per fuel molecule
BTDC Before top-dead center
CCmol Carbon monoxide concentration in exhaust % molar
C02mol Carbon dioxide concentration in exhaust % molar
EXHtot Exhaust flow rate g/hr
FUELppM Sulfur content in fuel ppm by wt.
H2 0air Water in ambient air % by wt.
H2Ocombmol Water in exhaust due to combustion % molar

JTH2Ocombwt Water in exhaust due to combustion %/by wt.
H20exh Water in exhaust % by wt.
H20wf Fraction of water in air
Relative air to fel ratio
LUBEs Sulfur content in lubricating oil % by wt.
MWexh Molecular weight of exhaust gases g/g-mole

NOC Normalized oil consumption
OC Oil consumption g/hr
Pa Density of air grams/m 3
Pf Density of fuel grams/cc

Sexh Sulfur dioxide flow rate in exhaust g/hr
SAMPvolt Detector output for oc reading volt
SOC Specific oil consumption g/kw-hr
SPANppm Span gas ppm by weight ppm by wt.
SPANvolt Output voltage from the detector due to span gas volt
T Temperature Celsius
13
va Air volumetric flow rate m3 /hr
vf Fuel volumetric flow rate cc/hr
y Relative hydrogen to carbon ratio
14
Chapter 1 MOTIVATION
1.1 Introduction
During the last several years oil consumption (OC) in diesel engines for both light
and heavy duty uses has been of major concern due to the increase of regulatory emission
controls. Recently, the EPA has set strict standards for heavy-duty diesel particulate
emissions. The 1994 maximum particulate rate allowable was restricted to 0.10 g/bhp-hr,
and the standard for 1996 for urban buses is 0.05 g/bhp-hr [1]. This drastic decrease in
particulate rate allowable for diesels has started a new age in the development of clean
diesels.
Oil transport and consumption is one such research area in which engine designers
can help to improve future particulate rates. Laurence reported that oil can contribute
between 24-33% of the soluble organic fraction in a small light-duty diesel engine [2].
Yoshida et. al. have reported that recent improvements in combustion efficiency have
resulted in the soluble organic fraction contributing up to 50% of the total particulate rates
in diesel engines [3]. This ultimately results in the OC contributing to between 12-22% of
the particulate rate. With such a significant portion of the overall particulate rate, OC
needs to be further studied in order to meet regulatory standards into the next century.
1.2 Previous Work
Oil consumption has been measured using many different methods over the past
thirty years. Some of these OC data show large variability and unpredictable behavior.
One such study was conducted by Lusted [4]. Lusted studied the OC behavior in a
production four cylinder spark ignition engine using a tritium radiotracer oil consumption
measurement system. In his initial experiments using unpinned rings and measuring the
OC of just one cylinder, he found a significantamount of variability, greater than 20 %, in
his data at one specific speed and load. At other speeds his OC measurements had much
15
lower variability. Another study conducted by Hartman in 1990 measured the OC of a
single cylinder Kubota diesel engine [5]. His results also showed large unexplained
variability at one speed and load. Others have mentioned that real-time OC results, using
sulfur dioxide tracing techniques of heavy duty diesels at steady state conditions,
sometimes have a load and speed where the OC is either variable from day to day or
abnormally high [6]. These variability effects introduce questions as to the source and
cause of such differences in measurements. This critical speed variability behavior
warrants further investigation, without which, a significantreduction of particulate rate
may be difficult to achieve.
1.3 Oil Consumption Sources
In diesel engines there are four major oil consumption paths. Hill et al., identifies
these as listed below [7]:
1. Piston-ring-liner system
2. Overhead valve seals
3. Crankcase Ventilation system
4. Oil pan, crankcase, turbocharger, and other component gasket and seals
In certain diesel engines where the crankcase ventilation is not coupled to the air inlet, the
two oil consumption paths which are involved directly in the air transport and combustion
process are the piston-ring-liner system and the overhead valve seals. Also, for
turbocharged engines, some oil can be combusted due to turbo seal leakage. Hill et. al.
have reported that in some diesels the valve system can contribute as much as 50 percent
of the overall oil consumption [7]. In another study, Ariga et al. reported that in a
particular gasoline engine, the overhead valves contributed as much as 80 percent of the
entire engine oil consumption at one speed and load [8]. This implies that the two main
mechanisms of combustion chamber derived OC are significant and need to be investigated
individually in order to better understand overall engine oil consumption.
16
1.4 Oil Consumption Measurement Techniques
Several different OC measurement techniques have been developed in the past 25

years. They best can be described in two separate categories, time-averaged and real-time
methods. The first being an average OC based on a specific sampling time interval. Real-
time measurements provide instantaneous OC indications, however they are subject to the
thermal and flow response times of the exhaust sampling systems.
Steady-state methods have been used extensively during the past two decades.
Two of the most widely used methods are the weight method and the tritium tracer type
method. The weight method is a simple method of oil consumption by just running the
engine for 3 or more hours and measuring the loss of oil. This technique requires large
running times, an oil leak-free engine, and an accurate measurement device. Another
inherent problem has been discussed by Butler, et al. [9]. They mention that in certain
cases, this type of methodology is flawed when fUel is diluted in the oil during operation.
Another steady-state method that has been used often is the radioactive tracer
technique. This method uses a radioactive element in the oil as a tracer in the exhaust
stream. By analyzing the radioactive content of the exhaust stream over a specified
period, the steady-state OC can be determined. One drawback of this method is in the
care and operation of the engine and surrounding equipment [10]. Also because tritium is
hard to detect, it is difficult to measure instantaneously. This makes detecting variations in
OC difficult.
Real-time methods have also been developed to measure instantaneous OC ill both
diesels and gasoline type engines. These methods trace the sulfur content, which is found
in the lubricating oil, in the engine exhaust. By measuring the sulfur dioxide content in the
exhaust, the OC can be evaluated. These methods use many different types of detectors.
Several studies have used both flame photometric detectors and coulemetric cells [11],
[12]. Others have used fluorescence type detectors [8]. These types of detectors can
measure both steady-state and transient OC accurately without the precautions of using
radioactive tracers. Also, results can be seen while the engine is in operation, hence,
enhancing the analysis of such variations in oil consumption on short time scales.
17
These sulfur dioxide tracing methods usually require low sulfur fuels and specially
formulated high sulfur lubricating oils. The fuel needs to be low in sulfur to not introduce
sulfur dioxide derived from the fuel in the exhaust. The oils need to be specially
formulated high sulfur oils having uniform sulfur distribution and high enough sulfur levels
for the detection equipment. With better instrumentation currently available, commercial
oils may be successfully implemented in these methods.
1.5 Objectives
The background above shows that a more in depth study of the OC mechanism is
required. By studying individual OC mechanisms while using a real-time sulfur dioxide
tracing technique, the cause of data variabilityand the effect of engine design parameters
may be determined. The current research addresses these issues in a single cylinder diesel
using the following set of objectives:
1. Establish and apply the procedures for a real-time sulfur dioxide oil
consumption technique for use in a set of detailed oil consumption
studies subsequent to this current work.
2. Study ring-pack oil consumption variations, if any, at steady-state
engine operating conditions.
a. Analyze the steady-state oil consumption and variability with
standard ring configuration.
b. Analyze the steady-state oil consumption and variability with a
lower tensioned oil control ring.
3. Investigate the use of a commercially available lubricating oil and its
performance in the sulfur dioxide testing method.
18
Chapter 2 EXPERIMENTAL APPARATUS
2.1 ENGINE
The engine used in this experimental study was a Ricardo Hydra single-cylinder
direct injection naturally aspirated diesel engine configured by G. Cussons Ltd. to a
research test cell set-up with oil and water pumps mounted separately from the engine.
The load control and starting of the engine were accomplished with the use of a
Dynamatic dynamometer and variable speed coupling mounted on the engine driveshaft.
The specifics of the engine are found in Table 2-1 below.
MANUFACTURER: G. CUSSONS LTD.

MODEL: HYDRA RESEARCH ENGINE
NUMBER OF CYLINDERS: 1
DISPLACEMENT: 0.45 L
BORE: 80.3 MM
STROKE: 88.9 MM
MAXIMUM SPEED: 4500 RPM
COMPRESSION RATIO: 20:1
INJECTION: DIRECT
Table 2-1 Engine Characteristics
The engine set-up included separate controllers to control the fuel flow, load, and engine
speed independently. This allowed many options while running the engine, including
accurately repeating a given operating condition.
Due to the limitations of the dynamometer, the maximum speed of the engine in
the current design was 3400 rpm. The manufacturer suggests that the engine's power be
limited by the exhaust smoke. Their maximum power output was 8 kW at a smoke level
of 4 Bosch units at 4500 rpm. The fuel injection timing could also be adjusted in order to
optimize the combustion and hence the performance of the engine.
Several k-type thermocouples were place in different engine locations. Table 2-2
below describes the different thermocouples.
19
THERMOCOUPLE DESCRIPTION
NUMBER
1 VALVE OIL
2 FUEL IN
3 OIL INTO BLOCK
4 COOLANT TANK
5 EXHAUST MANIFOLD
6 AIR INTO HEAD
7 EXHAUST MIXING TANK
8 OIL OUT OF SUMP
9 AIR INTO HEAD
10 OIL INTO BLOCK
11 COOLANT TANK
12 DYNO COOLANT
13 COOLANT INTO ENGINE
14 CYLINDER COOLANT OUT
15 HEAD COOLANT OUT
16 TEST CELL
Table 2-2 Engine Thermocouples
The engine crankcase was equipped with a crankcase blow-by system. This was a
5/8 inch exhaust hose connecting the engine crankcase to the test cell's exhaust trench.
The exhaust system was set up in accordance to the study done by Laurence [2].
It included a "marine" style exhaust system with aqueous injection to cool the exhaust.
This exhaust system was placed well downstream of the oil consumption sampling location
to not effect the oil consumption results. Beyond the aqueous injection apparatus the
exhaust gases were separated in a mixing tank and exhausted into the test cell exhaust
trench. The exhaust system also was fitted with a BG- 1 Micro Dilution Tunnel for the
availability of particulate measurement. Figure 2-1 below shows the "marine" type
exhaust system.
20
- ENGINE
E:
- MIXING
TANK
TO
PARTICULE
TO GAS
FILTER CART
I - THERMOC
WATER IN
_ VALVE EXHAUST
COOLER
SEPARATION
TANK
EXHAUST
UT
I I
Figure 2-1 Exhaust System
The engine was also equipped with the capabilityto isolate the overhead valve
train oil with that of the engine sump. This was easily accomplished due to the engine's
external stainless lines supplying oil to the overhead valve system. The engine test bed
was modified with the addition of a 1/3 horsepower motor connected to a 0.5 gallon per
minute positive displacement pump. Beyond the pump, a relief valve was installed and set
21
to 50 psig. An oil cooler was also installed to control the temperature of the valve oil.
Figure 2-2 below shows the details of the external overhead valve oil system.
TCH
JRN
PLASTIC
1/2 GPM TUBING
1/3 HP SPUR GEAR
1750 RPM PUMP \ 10 MICRON
120 VAC FILTER
MOTOR
Figure 2-2 Valve Oil System
2.2 Oil Consumption Apparatus
The oil consumption of the diesel engine was measured with a pyrofluorescence
sulfuirdioxide testing apparatus developed by Cummins Engine Company, Inc. Figure 2-3
shows the schematic of the instrumentation installation.
22
EX U ENGINE
EXHAUST
1/4 " S/S

TUBING
SPAN/ZERO
AIR CELL AIR
[M_ THE] FURNACE

1000C
3- SOL
[- - FIL
- HEA
LIN
AMPLE
Y-PASS
DRYER
R
OZONEA
TPUT TO
CHART
CORDER
OUTPUT TO
S02
COMPUTER
ANALYZER
Figure 2-3 OC Sampling System
The general principal of the system was to trace the sulfur in the oil through the
combustion process and into the exhaust stream. This was accomplished by using a zero
23
sulfur diesel fuel and with lubrication oils containing sulfur. By analyzing the sulfur
content in the exhaust, the oil consumption could be obtained.
The sampling line was placed in the exhaust manifold near the engine head. The
sample exhaust gas went through heated lines to maintain a high exhaust temperature into
a group of solenoid valves controlling whether the exhaust, span gas, zero gas, or ambient
test cell air would pass through the system. Next the sample passed into a furnace which
operated at 1000°C. Inside the furnace, a quartz combustion tube completed the
oxidation of the unburned fuel and oil. Upon exiting the furnace, the sample was filtered
with a 60 micron filter. This extracted the remaining particulate out of the sample. It then
passed through two heated lines into a membrane dryer which removed the 3-12% water
in the exhaust directly from the vapor phase without condensing the moisture first. The
sample was then mixed with ozone to oxidize the NO to NO2. Finally, the sample's sulfur
dioxide concentration was measured in molar concentration in a pulsed fluorescent
ambient SO2 detector. This concentration reading had linear dc voltage outputs which
were tied into a chart recorder and a computer data acquisition system.
The outputs were real-time measurements of the sulfur dioxide concentration
which was directly related to the OC in the engine. The details of the conversion is
discussed in Chapter 4. A chart recorder recorded the results instantaneously during
testing. Its speed was set at 30 cm/hr. This gave visual results while the engine was
operating.
The output of the sulfur dioxide and several other engine operational variables
were recorded directly on a computer data acquisition system. This systein included four
dc voltage outputs from the engine testing cell. These included; air flow, fuel flow,
torque, and sulfur dioxide voltage outputs. These signals were input into a computer data
acquisition system using a software system, Global Lab. The data collection was then
saved to a disk in order to make the appropriate conversions in the actual data analysis.
A laboratory manual [13], has been prepared for this work that contains a more
complete description of the OC system and operation. Dr. Victor Wong has a copy of this
manual.
24
Chapter 3 EXPERIMENTATION
3.1 General Overview Of The Testing Procedure
This chapter presents the testing procedure for the oil consumption data. The
actual testing was categorized into three matrices, A, B, and C. All three matrices used
zero sulfur diesel and oil containing sulfur. The A matrix was accomplished using a
standard oil control ring tension. Matrix B was completed with a lower tensioned oil
control ring. The A and B Matrices had only ring-pack derived oil consumption. Matrices
A and B were completed using high sulfur oil, while Matrix C had oil at a normal level of
sulfur content. A summary of all three test matrices is given below in Table 3-1.
LATRIX OC OIL TYPE OIL OIL BREAK-

DERIVATION CONTROL IN
RING (HRS)
TENSION
A RING-PACK LUBRIZOL 30W STANDARD 50
B RING-PACK LUBRIZOL 30W LC WER 95
C RING-PACK & SHELL 10W-30 STANDARD 25
VALVES
Table 3-1 Overview of Testing
The table shows the order in which testing was completed and several of the
variables in the testing. Matrix A and most of Matrix B was completed in conjunction
with particulate testing completed by Eric Ford [14]. The engine itself was overhauled
twice during the testing in order to change out the oil control ring. During these periods
the engine was re-assembled as close to the original configuration as possible. The one
exception was with respect to the hydrocarbon build-up on the cylinder liner. These
deposits had to be dissolved with acetone to facilitate the piston assemblyremoval. Each
of the testing variables will be covered in more detail in Sections 3.2 through 3.4.
25
3.2 Testing Procedure For The Standard Oil Control Ring Tension
3.21 Test Matrix A
Table 3-2 below shows the details of Matrix A.
NAME SPEED LOAD (bmep) # OF TESTS

A 1200 LOW (1.4) 4
B 1200 HIGH (4.5) 4
C 2400 LOW (1.4) 4
D 2400 HIGH (4.5) 4
E 3300 LOW (1.4) 4
F 3300 HIGH (4.5) 4
OIL CONTROL
RING TENSION: "STANDARD"
ENGINE OIL: LUBRIZOL 30W
OC DERIVED: RING-PACK
I [ I I ll Ill I I ll
Table 3-2 Test Matrix A
This matrix included three different speeds, two different loads, and four separate tests for
each load and speed resulting in a total of 24 tests. A low sulfur oil was used for the
overhead valve lubrication and a high sulfur oil was used to lubricate the crankshaft and
piston assemblies. Thas was done to analyze the OC derived mostly from the ring-pack.
The details of the variables can be found in Section 3.23.
3.22 Procedure
In order to ensure repeatable results, the oil was first "broken-in" for 50 hours.
This was to ensure that the oil volatility and viscosity levels were at a steady-state level.
After the initial break-in period, testing followed with up to 6 tests completed in a single
day.
A typical day of testing was started first with the OC system purging and
calibration. A dry zero sulfur gas was input into the system for approximately 45 minutes
in order to clean out the system and establish the baseline zero setting. After the initial
26
zeroing, a standard span gas was sent through the sulfur dioxide detector to establish a
span setting. The span setting was adjustable via a span manual pot and also by a photo-
multiplier setting in the sulfur dioxide detector housing. The span setting was then
stabilized according to the manufacturer's specifications for a minimum of 6 minutes.
During this time, the linear chart recorder setting and the voltage output to the computer
were logged in a data sheet contained in Appendix A, Figure A-1.
After the initial calibration procedure, the engine was started and warmed up at
idle, 1200 rpm, for approximately 10 minutes. After engine warm-up, the engine was run
at the maximum speed and load, 3300 rpm-4.5 bmep, and the oil consumption sample line
was opened. This was done in order to burn any residual fuel and oil in the exhaust lines
and to allow the oil consumption apparatus time to "settle out". This period was
approximately 30 minutes. After this initial procedure, the testing was started.
Each day the first test point was Condition F, the maximum speed and load. Each
test point was approximately 45 minutes in duration and included data collection for both
the oil consumption data mentioned in this study and also data for a simultaneous study by
Eric Ford for particulate rate, emissions, and aqueous injection data. The following list
describes the procedure for a typical test:
Time: 0 Minutes: Steady-state engine speed and load obtained.
Time: 10 Minutes: Particulate samples taken for 2-5 minute duration.
Time: 20 Minutes: Oil consumption computer data acquisition started,
engine control informationrecorded, and aqueous injection water
samples taken.
Time: 30 Minutes: Exhaust emissions recorded for both before and after
aqueous injection.
Time: 40-50 Minutes: End of test, change load and speed.
The first ten minutes was used for the engine to reach steady-state operation. The
temperatures described in section 2.1 were watched during this time. Most of the
temperatures of the engine reached a steady reading within 5 minutes with the exception
of the oil temperatures. The temperature of the oil out of the engine sump and the oil out
the oil pump took from 5-15 minutes to reach a steady temperature.
At the ten minute point, the particulate sampling was completed. This involved a
computer controlled sampling system which first purged the sample line for 60 seconds
27
and then took an exhaust sample for between 2-4 minutes. At the lower speeds, the
engine load decreased during the filter purging of the particulate testing system due to an
increase in back pressure. This did not ultimately affect the filter or subsequent OC
testing.
After 20 minutes the oil consumption data collection was started using the
computer data acquisition system described in Chapter 2. The data acquisition lasted for
10 minutes with data collected from the four channels with a frequency of 0.2 samples per
second, or a sample collection every 5 seconds. Simultaneously, engine temperatures, etc.
were recorded onto a data sheet contained in Appendix A, Figure A-2.
At the 30 minute mark, the OC computer data was output to a disk and the
exhaust emissions were recorded.
This same procedure was completed for all the tests in Matrix A and the chart
recorder was on throughout each day of testing. To facilitate the testing and to make the
steady-state transition time minimal,the following order of testing was followed; F, E, B,
A, D, C. Also, after each low load test, the engine was run for 5 minutes at the maximum
speed and load. At the end of the day of testing, the calibration procedure was repeated in
order to ensure accurate results throughout the day.
3.23 Matrix A Engine Specifics, Controls, and Variables
During the testing, the engine was run with a ultra low sulfur diesel so that the fuel
would not contribute to the overall sulfur dioxide concentration. Table 3-3 below gives a
summary of the diesel fuel characteristics.
28
API GRAVITY 39.8
FLASH POINT (F) 156
POUR POINT (F) -5.0
CLOUD POINT (°F) -10.0
VISCOSITY (Cs @40 °C) 2.7
SULFUR (weight ppm) 0.1
HYDROGEN-CARBON RATIO 1.88
HEAT OF COMB (Btu/lb) 18522.0
PARTICULATE MATTER (mg/l) 5.06
CEIANE INDEX 55
CETANE NUMBER 42
Table 3-3 Diesel Fuel Characteristics
Throughout this matrix, the overhead valve oil system was separated from the
main oil system. The characteristics of both oils are summarized in Table 3-4.
LUBRICATION BRAND TYPE SULFUR CONTENT

(% by weight)
VALVE SHELL FIRE AND SAE 10W-30 0.279
ICE ALL SEASON
MAIN ENGINE LUBRIZOL HIGH SAE 30W 1.27
SULFUR
Table 3-4 Oil Characteristics
The ring-pack configuration was the standard as per the manufacturer's

specifications. The first ring or compression ring was chrome plated and slightly rounded
in shape. The second ring or scraper ring was beveled. The third ring or oil control ring
was chrome plated with two rails and a separate coil spring for tension control. Although
the ring characteristics were not measured directly, Table 3-5 below shows the
manufacturer's specifications.
29
RING DIAMETER (mm) COLD GAP (mm) TENSION (N)
COMPRESSION 80.25 0.43 9.3
SCRAPER 80.25 0.43 8.2
OIL CONTROL 80.25 0.51 53.8
Table 3-5 Piston Ring-Pack Characteristics
The importance of controls cannot be overlooked. The most important control

feature was the repeatability of the engine temperatures described in section 2.1. Of these,
the oil temperatures were continually monitored to ensure the engine had reached a
steady-state operation. However, there was no external heating or cooling of the oil to
specific pre-set temperatures.
Matrix A had two variables, speed and load. The brake mean effective pressure
(bmep) was kept constant throughout the speed range of the testing. Table 3-6 below
shows the target load, bmep, fuel flow rate, and exhaust temperature for the testing.
SPEED ENGINE BMEP FUEL FLOW INJECTION EXHAUST

(rpm) LOAD (bar) RATE TIMING TEMP
(cc/min) °BTDC (C)
1200 LOW 1.4 7.0 9 235
1200 HIGH 4.5 15.3 9 445
2400 LOW 1.4 14.8 14 280
2400 HIGH 4.5 28.5 14 480
3300 LOW 1.4 24.3 15.5 375
3300 HIGH 4.5 46.1 15.5 610
Table 3-6 Matrix A Variable Specifics
3.3 Testing Procedure For The Less Tensioned Oil Control Ring
3.31 Test Matrix B
'Fable 3-7 below shows the details of Matrix B.
30
I 1200 LOW (1.4) 4
J 1200 HIGH(4.5) 4
K 3300 LOW (1.4 4
L 3300 HIGH (4.5) 4
T 1200 HIGH (4.5 3
U_ 3300 HIGH (4.5) 3
OIL CONTROL
RING TENSION: LOWER
ENGINE OIL: LUBRIZOL 30W
OC DERIVED: RING-PACK
Table 3-7 Test Matrix
This matrix was done with similar parameters to that of Matrix A. Again, the
overhead valve system was separated. The major change was that a modified oil control
ring with 45 % less tension was substituted for the original in the ring-pack.
3.32 Procedure
The procedure for tests I through L were done as per the procedure addressed in
Section 3.22. The tests, T and U, were done to study the long term variability of the data
and were not completed in conjunctionwith particulate or emission testing. Their
procedures are addressed below.
For these tests, the calibration procedures were repeated as before. Each test had
a duration of 60 minutes outlined as follows:
Time: 0 Minutes: Steady-state engine speed and load obtained.

Time: 15 Minutes: Oil consumption computer data acquisition started.
Time: 30 Minutes: Engine control information recorded.
Time: 60 Minutes: End of test, change load and speed.
The first fifteen minutes was used for the engine to reach steady-state operation.
The temperatures described in Section 2.1 were watched during this time. Most of the
temperatures of the engine reached a steady reading within 5 minutes with the exception
31
of the oil temperatures. The temperature of the oil out of the engine sump and the oil out
the oil pump took from 5-15 minutes to reach a steady temperature.
After 15 minutes the oil consumption data collection was started using a computer
data acquisition system described in Chapter 2. The data acquisition lasted for 45 minutes
with data collected from the four channels with a frequency of 0.2 samples per second, or
a sample collection every 5 seconds.
At the 30 minute point, the engine control parameters of speed, load, etc. were
recorded onto a data sheet contained in Appendix A, Figure A-2.
At the 60 minute mark, the OC computer data was output to a disk and the
exhaust emissions were recorded.
This same procedure was completed for all tests T and U in Matrix B and the chart
recorder was on throughout each day of testing. At the end of the day of testing, the
calibration procedure was repeated in order to ensure accurate results throughout the day.
3.33 Matrix B Engine Specifics, Controls, and Variables
All of the engine specifics, controls, and variables were the same as Matrix A with
the exception of the ring-pack oil control ring tension. A new oil control ring was
installed with a lower tension as shown in Table 3-8.
RING DIAMETER (mm)" COLD GAP (mm) TENSION (N)

OIL CONTROL 80.25 0.51 30.3
Table 3-8 Modified Oil Control Ring Characteristics
3.4 Testing Procedure For The CommerciallyAvailable Oil

3.41 Test Matrix C
Table 3-9 below shows the details of Matrix C.
32
REGA 1200 LOW (1.4) 4
REGB 1200 HIGH (4.5) 4
REGF 3300 HIGH (4.5) 4
OIL CONTROL
RING TENSION: "STANDARD"
ENGINE OIL: SHELL 10W-30
OC DERIVED: RING-PACK & VALVES
Table 3-9 Test Matrix C
This matrix was completed with the engine oil system in its standard configuration with
the sump oil also being distributed to the overhead valves.
3.42 Procedure
The procedure for all of these tests were the same as the procedure from Matrix B
for the tests, T and U. Prior to the testing, the engine was run at the maximum speed and
load for 25 hours in order to break-in the oil prior to testing.
3.43 Matrix C Engine Specifics, Controls, and Variables
All of the engine specifics, controls, and variables were the same as Matrix A with
the exception of the engine oil. Instead of using a high sulfur oil, a commercially available
oil was used. The oil was a commerciallyavailableShell oil. Its characteristics are listed
below in Table 3-10.
LUBRICATION BRAND TYPE SULFUR CONTENT

(% by weight)
ENTIRE SHELL SAE 10W-30 0.8
ENGINE ROTELLA T®
Table 3-10 Commercially Available Oil Characteristics
33
Chapter 4 THEORY
4.1 General
The output of the analyzer is in terms of sulfur dioxide volumetric concentration in

the exhaust. This section analyzes how to convert the sulfur dioxide concentrations into
useable oil consumption results.
The first important factor in the analysisof the data is that the concentration of the
sulfur dioxide is detected in a dry exhaust gas stream. The water content of the wet
exhaust must be known in order to account for this lost water. By using dry gaseous
emissions, Heywood [15], offers an equation to determine the molar concentration of
water in the exhaust derived from the combustion process. Equation 4-1 describes this
equation.
H2 0combmol= 0.5y CO2m! + COm. (4-1)

COmo]
3.8(CO2mo,) + 1
where:
H20combmol = water in exhaust due to combustion (% molar)

C02mOI = carbon dioxide concentration in exhaust (% molar)
COmo
1 = carbon monoxide concentration in exhaust (% molar)
y = relative hydrogen to carbon ratio
Now, given a fuel molecule comprised of CaHband the fuel and air flow rates and
densities, a simple equation can be derived which calculates the relative air to fuel ratio, X.
34
(va )(Pa)
(4-2)
34.56(4 + y)
12.011+1.008y
where:
va = volumetric air flow rate (m3 /hr)

Vf = volumetric fuel flow rate (cc/min)
Pa = density of air (g/m 3 )
pf = density of fuel (g/cc)
y = relative hydrogen to carbon ratio
Using this relative air to fuel ratio, Laurence [2], gives the following equation to
calculate the molecular weight of the exhaust.
MW
MW= 6a+18+ 16([2(a +)-(2a + ))+ 106.25A(a+ 4)
a+ 2+ 0.5([2(a-+ (2a+ ) + 3.773(a + )
where:
MAWCh = molecular weight of exhaust (g/g-mole)

a = number of carbon atoms per fuel molecule
b = number of hydrogen atoms per fUel molecule
Next, usihigthe exhaust molecular weight, Equation 4-1 can be converted to

weight basis using the Equation 4-4.
m bmoo l
H mm=1 8 H2 0
Mbm (4-4)
where:
H2 0 ,mbwt = water in exhaust due to combustion (% by weight)
With a known relative humidity and temperature, psychrometric charts can be used
to find the weight fraction of water in the air. This can ultimately lead to the water
35
content in the air. Using this percentage and that of the combustion derived water, the
total water content of the exhaust can be found.
H20air (H 2 Ow)(PaVa)(100) (45)

EXHtot,
H2 h = H2 0 combwt+ H20air (4-6)

where:
H2 Owf = fraction of water in the air

EXHtot = exhaust flow rate (g/hr)
H2 Oar = water in the air (% by weight)
H 2 0exh = total water in exhaust (% by weight)
Using an output voltage from the detector for a given test, the voltage and
concentration of sulfur dioxide in a span gas, the total sulfur dioxide flow rate in the dry
exhaust can be determined.
Sah = (SAMPv,):t
(EXH,)(PANPP) ()
S.b Jg (4-7)
,S-o
SPANTL
2x106
2 x 106
where:
SPANpro = span gas sulfur dioxide concentration (ppm by weight)

SPANvolt = output voltage of detector using span gas (volt)
SAMPVooI: detector output of current test (volt)
Sexh = sulfur dioxide exhaust flow rate (g/hr)
From this sulfur dioxide flow rate, the oil consumption can be found using the
equation below. This equation includes the correction of the fuel sulfur content and also
the ambient air concentration. It also corrects for the water concentration in the exhaust
gas stream.
36
OC=s [(pf)(Vf)(FUEL)
Lx 106
I (AIRo,,) ()(V 8 )(PAN
(SPANVO,) 2 x 106
)lY 1(
LUBE
)')i1]
(4-8)
where:
FUE,Lp, = sulfur content in fuel (ppm by weight)

AIRvolt = output voltage of detector using cell air (volt)
LUBE, = sulfur content of lube oil (% by weight)
OC = oil consumption of engine (g/hr)
4.2 Matrix Application
The equations above were used to etermine both the average oil consumption and
also the instantaneous oil consumption traces. The following shows how these equations
were applied to the testing.
After testing, the computer outputs of sulfur dioxide voltage, air flow, fuel flow,
and torque as well as gaseous emissions and calibration data were transferred into a
spreadsheet. Using the average values of fuel flow and air flow, the water content of the
exhaust was computed using Equations 4-1 through 4-6. Using this value and each
sample's real-time voltage, the sulfur flow rate in the exhaust was determined using
Equation 4-7. Then, using Equation 4-8, the instantaneous oil consumption trace values
were determined. Averaging these values as well as the torque outputs, each test's
average OC and SOC were determined.
37
Chapter 5 RESULTS AND ANALYSIS
5.1 Results For The Standard Oil Control Ring

5.11 Matrix A Steady-State Oil Consumption
Figure B-1 in Appendix B shows the steady-state results that were calculated in
accordance to the equations in Chapter 4. Figures B-2 and B-3 show the OC results in
graphic form. Table 5-1 below shows the specific oil consumption (SOC), time averaged
over the 10 minutes testing cycle, for each of the four tests, their sample standard
deviations, and their coefficients of variation.
CONDITION AVERAGE SAMPLE COEFFICIENT

SOC STANDARD OF
(g/kw-hr) DEVIATION VARIATION
(g/kw-hr) (%)
A-1200 LOW 0.220 0.012 5.61
B-1200 HIGH 0.199 0.035 17.65
C-2400 LOW 1.147 0.224 19.51
E-3300 LOW 0.752 0.095 12.68
F-3300 HIGH 0.440 0.006 1.28
Table 5-1 Summary of Matrix A Steady-State SOC Results
The above results do not include those of Condition D from the original test
matrix. At this operating point, 2400 rpm-4.5bmep, the oil consumption was higher than
that which could be measured by the oil consumption apparatus. The SOC at this point
appears to be higher than approximately 1.34 g/kw-hr. Condition C also had relatively
high SOC values with high variability. This suggests that there may be a dynamic effect in
the ring-pack at the 2400 rpm speed that causes higher SOC.
The data shows that at Conditions A and F, there is excellent repeatability in SOC
from day to day testing. The trends show that the SOC at a given speed is lower at the
higher load, which is what would be expected. Also, the trends indicate that the SOC is
increasing from low speed to a high speed with a peak in SOC at the mid speed range
typical of many diesels.
38
5.12 Variability Of Matrix A During Steady-State Operation
Appendix C shows graphs of the computer real-time results of each operating

condition over the ten minute testing cycle. The OC values were normalized with respect
to the average OC values of all four tests. This normalization makes it easier to compare
different conditions and matrice.
The results show similar trends to those found in Section 5.11 above. For the
most part, the graphs all show smallvariability patterns, deviation less than 10 percent of
each test's mean OC value. The small "peaks and valleys" during many of the tests may
represent small fuel and load variations that are frequently observed in single cylinder
engines. Conditions A, E, and F show this pattern with no large OC variations in each test
or from test to test. Conditions B and C show some larger variability in the average OC
values from test to test.
Figure C-2 shows the results of Condition B. There is a definite change in OC
from test to test, and this OC level is not a function of the actual load on the engine. Test
B4 has a relatively high OC, but with the lowest average load for the testing duration.
Test B 1 shows a OC trace with larger than 10 percent variability in OC. The periods of
these are between 1 and 3 minutes in duration. Again this may be a dynamic effect in the
ring-pack.
Figure C-3 shows the results of Condition C. This data is also variable from test
to test and is not a function of the actual average load. Another interesting result can be
seen in Test C3. Near the end of the test, the OC increases by 50 percent. Studying the
linear chart recording of this test, Figure C-6 in Appendix C, shows interesting trends.
When the engine was first run at Condition C, an OC "plateau," constant OC period,
occurred. This constant OC continued for approximately 12 minutes, then increased by 25
percent to a plateau lasting 15 minutes during which time the data for Test C3 was
recorded. This non-periodic phenomenon is not easily explainable. While these plateaus
were occurring, the engine load, fuel flow, etc. did not noticeably change.
39
The chart recording of Test C2, Figure C-7 in Appendix C, also shows similar type
results. At the beginning of Test C2, the OC was at a constant "plateau" for about 30
minutes during which the C2 results were recorded. At this point the OC instantly
increased by 100 % to a plateau lasting 15 minutes, at which time the OC trace instantly
changed back to the original OC level.
5.2 Results For The Less Tensioned Oil Control Ring

5.21 Matrix B Steady-State Oil Consumption
Figure D-1 in Appendix D shows the steady-state results that were calculated in
accordance to the equations in Chapter 4. Figures D-2 through D-5 show the OC and
SOC results in graphic form. Table 5-1 below shows the SOC results, time averaged over
the 10 minutes testing cycle, for each of the four tests, their sample standard deviations,
and their coefficients of variation.

SOC STANDARD OF
(g/kw-hr) DEVIATION VARLATION
kw-hr) (%)
I-1200 LOW 0.183 0.030 16.29
J-1200 HIGH 0.195 0.050 25.96
L-3300 HIGH 0.718 0.149 20.71
T-1200 HIGH 0.166 0.038 22.89
U-3300 HIGH 0.678 0.131 19.32
Table 5-2 Summary of Matrix B Steady-State SOC Results
The results do not include data from Condition K, 3300 low load, because during
this condition, the engine operation was unsteady. The load varied as much as 75 percent
and the engine sounded as if poor combustion was taking place. Also, the first set of tests
were not recorded due to problems with the OC apparatus.
These results show a large variability in the average SOC values as shown by the
high coefficients of variation. Unlike the results from Matrix A, Conditions I and L, 3300
40
rpm-4.5 bmep, even show large deviations from test to test. This is further analyzed in
Section 5.22.
5.22 Variability Of Matrix B During Steady-State Operation
This matrix included both the original ten minute OC tests as well as longer 45
minute duration tests, Conditions T and U. These results show high variability and
cyclical spikes of OC. Appendix E contains the graphs of the real-time OC results with
normalized OC values that were normalized with respect to the average OC values of all
four tests.
Conditions J3 and J4 in Figure E-2 both contain very similar plots of OC. In both
cases there is a "spike" in the OC plots during the middle of the tests with about 50%
hig-er instantaneous OC. Both spikes occur at approximately 24 minutes after the tests
were started. Test J3 also had two smaller spikes one 2 minutes prior to the large spike
and the other 4 minutes after the large spike.
Condition L also shows patterns of unstable OC during the 10 minute testing
period not seen in Matrix A. Test L2, Figure E-3, shows the same "plateau" effect as was
found during Matrix A, Condition C. This plateau shows an OC of about 58 % higher
than the steady-state period before or after with a duration of about 7 minutes. Condition
L5 shows a spike of oil consumption that was unmeasurable but greater than an OC of 6
g/hr.
Because of the large variability of data in Tests I, J, and L, longer term Tests, T
and U, were completed. These tests were completed as per Section 3.32. Condition T,
Figure E-4, shows a large spiking phenomenon during the 45 minute testing. Condition
T3 shows a large spike and a smaller spike with a 13 minute period. Condition T2 shows
two small spikes with a 4 minute period.
Figure E-5 shows very interesting results. Test U2 shows two spikes during the
testing with an instantaneous OC increase of 75% and 35%, respectively. These spikes
are separated by approximately 20 minutes. Test U3 shows some interesting spikes with a
repeatable pattern. This test shows two large spikes in OC both of which were larger than
41
6 g/hr in magnitude. Just prior to the testing another large spike also occurred. The
period between the first two is about 14 minutes and the second about 16 minutes in
duration. Both of these spikes led to a very cyclical pattern with several smaller spikes
following the large spikes.
Overall, this Matrix B shows much more variability in steady-state OC than Matrix
A. The spikes and plateaus may be a phenomena explained by the studies conducted by
Hiruma et al., and by Schneider et al. [16], [17]. They suggest that ring dynamics such as
ring lift off, rotation, and ring twist may cause significant changes in OC. The
instantaneous spikes may be caused by a slow ring rotation phenomena in the ring-pack.
When the ring gaps line up, or the top ring lifts off its the piston groove, a relatively large
amount of oil may be transported into the cylinder by reverse blow-by. The OC plateaus
may be caused by the ring position becoming steady at another position. This temporary
position may be ultimately effected by the transport of oil in the ring-pack. The transport
of oil may accumulate in the second land and then suddenly when reaching a critical point,
cause the ring positions to change and hence a different OC.
5.3 Results For The Commercially Available Oil

5.31 Matrix C Steady-State Oil Consumption
Figure F-1 in Appendix F shows the steady-state results which were calculated in
accordance to the equations in Chapter 4. Figures F-2 and F-3 show the OC and SOC
results in graphic form. Table 5-3 below shows SOC values, time averaged over 45
minutes for each condition, their sample standard deviations, and their coefficients of
variation.
42
SOC STANDARD OF
(g/kw-hr) DEVIATION VARIATION
(g/kw-hr (%)
REGA-1200 LOW 0.376 0.097 25.80
REGB-1200 HIGH 0.143 0.028 19.58
REGF-3300 HIGH 0.769 0.022 2.86
Table 5-3 Summary of Matrix C Steady-State SOC Results
The coefficients of variation show excellent repeatability for the Condition REGF.
The other two conditions show larger repeatability errors which may be due to variability
discussed in Section 5.32.
5.32 Variability Of Matrix C During Steady-State Operation
Figures G-1 through G-3 in Appendix G show the steady-state OC results for
Matrix C. These graphs show different variations in the real-time OC than those found in
Matrix A or B. With the exception of REGB4, Figure G-2, no OC spiking occurred. At
the same time though, there was more variability in each test as well as between tests for
conditions REGB and REGA than those same conditions in Matrix A. This may be due to
the OC of this matrix was contributed by both the valves and the ring-pack. This
difference may be in the variability of the valve real-time OC. All three figures show a
trend that by the end of the testing period, the OC levels of the tests tend to end with very
close values. This may show that the engine reached a final steady-state OC period at the
end of 1 hour of operation at a given speed and load, potentially caused by an usteady
valve OC.
43
5.4 Comparison Of Matrices A, B, C
5.41 Ring-Pack OC Evaluation
Figure H- 1 of Appendix H shows a comparison of Matrix A results to that of the

entire engine SOC results that were conducted prior to this testing. By knowing the entire
engine SOC and that of the ring-pack only, the valve oil contribution can be calculated
using the oil consumption values, the sulfur content of the oils, and simultaneous
equations. Table 5-4 below summarizes the comparison.
CONDITION VALVE OIL VALVE OIL

CONTRIBUTION TO CONTRIBUTION TO
ENTIRE SOC (%) MATRIX A (%)
1200 Rpm Low Load 56 22
1200 Rpm High Load 47 17
3300 Rpm High Load 42 14
Table 5-4 Valve Oil Contribution
The results show that in this engine, the valve oil contribution is relatively high. It
also shows that there is a small contribution of OC from the valves in Matrices A and B.
5.42 Effects Of The Less Tensioned Oil Control Ring
Figure H-2 shows the comparison of the average SOC results for matrix A to that
of Matrix B. The results are compared in Table 5-5.
CONDITION CHANGE IN SOC FROM STANDARD

TO LESS TENSIONED OIL CONTROL
RING (%)
1200 Rpm Low Load - 16.8
1200 Rpm High Load - 2.0
3300 Rpm High Load + 63.2
Table 5-5 Matrix A to B SOC Comparison
44
This table shows an interesting result. By decreasing the oil control ring tension by
approximately 45%, the SOC values increased at 3300 rpm and decrease at 1200 rpm.
The expected result would be for the SOC to increase over the entire speed and load range
due to lower tensioned oil control ring causing higher oil film thickness' on the liner. One
factor that they may have contributed to these results is that the engine was rebuilt after
Matrix A in order to change out the oil control ring. Deposits on the liner had to be
removed in order to facilitate the piston removal. This may have ultimately affected the
low speed results.
5.43 Comparison Of Commercially Available Oil To High Sulfur Oil
Figure H-3 shows the graphical representation of the regular oil results of Matrix
C to that of the high sulfur results of the entire engine SOC prior to the testing. Table 5-6
shows the comparison.
CONDITION DIFFERENCE IN SOC RESULTS

USING COMMERCIALLY
AVAILABLEOIL (%)
1200 Rpm Low Load 2.6
3300 Rpm High Load 15.2
Table 5-6 Matrix C to High Sulfur Oil SOC Comparison
These percentage differences are small comparing these two operating conditions.
This suggests that this oil may be suitable for testing at these two operating conditions. A
source of error in comparing these two tests may be in the oil used. The "regular" oil was
a multi-grade 10W-30 oil while the high sulfur oil was a single grade 30W oil.
45
Chapter 6 CONCLUSIONS
1) The real-time sulfur dioxide oil consumption technique described in this work, provides
an excellent method to better study oil consumption in this single cylinder diesel
engine.
Oil Consumption Variationsin Time at Steady-StateEngine OperatingConditions:
2) Standard Oil Control Ring Tension: For this engine running at a steady-state speed and
load, several operating conditions were observed where the oil consumption values do
not remain constant. This unsteady oil consumption phenomenon occurred at 2400
rpm. At 2400 rpm-4.5 bmep, unusually high oil consumption was observed. At light
load, 1.4 bmep, the oil consumption was not steady, showing plateaus of oil
consumption lasting 15-30 minutes in duration with increases or decreases in oil
consumption between 25-100 %, even though the engine operating parameters were
kept steady. These plateaus may be caused by the axial and rotational dynamics of the
piston rings. This large variability of steady-state oil consumption data agrees with an
earlier study of a spark ignition engine [4].
3) Less Tensioned Oil Control Ring: For most engine operating conditions, unsteady oil
consumption behavior was present. These unsteady oil consumption characteristics
were different than those observed for the standard oil control ring tension tests, with
the exception of one condition. In contrast to the 15-30 minute plateaus, large spikes
in the real-time oil consumption measurementswere observed. The spikes are a rapid
rise and fall of oil consumption. At 1200 rpm-4.5 bmep, the time widths of these
spikes in oil consumption were on the order between 2-13 minutes. At 3300-4.5
bmep, the time widths were longer with values between 10-20 minutes. The
instantaneous spikes may be caused by a slow ring rotation phenomenon in the ring-
pack. Because of piston side motion, effective ring-gap sizes, and thus piston land
pressures and reverse blow-by, OC may be varying with ring orientation. When the
46
ring gaps line up in positions where high reverse blow-by occurs, or the top ring lifts
off the piston ring groove, a relatively large amount of oil may be transported into the
cylinder by reverse blow-by gases from the second land.
The Effect of the Oil ControlRing Tension on Time-AveragedSpecific Oil

Consumption:
4) The effect of decreasing the oil control ring tension by 45 % was to increase the
specific oil consumption at high engine speed, specifically at 3300 rpm-4.5 bmep. At
the low engine speed, 1200 rpm, the change in oil control ring tension did not result in
an appreciable change in specific oil consumption. In fact, at 1200 rpm-1.4 bmep, the
specific oil consumption decreased by 16 %, which is within the variability for this
series of tests.
Other Conclusions:
5) This engine's overhead valve system contributes between 42-56 % of the total specific
oil consumption related to the power-cylinder.
6) The sulfur dioxide diagnostics system was sensitive enough to measure sulfur dioxide
concentrations, hence oil consumption, using a commercially available 10OW-30
lubricating oil containing 0.8 % sulfUr by weight.
47
Chapter 7 RECOMMENDATIONS
1) This work has shown some very interesting and stimulating oil consumption behavior.
A more comprehensive study of the oil consumption mechanisms is strongly
recommended. By measuring other engine operating parameters such as ring rotation,
oil film thickness, etc. simultaneously,a better understanding of oil consumption
mechanisms might be obtained.
2) The underlying causes of the Uargevariability of oil consumption at 2400 rpm needs to
be examined in greater detail.
3) The feasibility of using commercially available oil needs to be fUrther explored in this
sulfur dioxide based oil consumption measurement system. This would allow for
greater flexibility in assessing engine oil consumption using different lubricating oils.
48
REFERENCES
1. Johnson, J. H., Bagely, S. T., Gratz, L. D., and Leddy, D. G., "A Review of Diesel
Particulate Control Technology and Emissions Effects-1992 Horning Memorial
Award Lecture," SAE Paper 940233, 1994.
2. Laurence, R. B., "The Effects of Lubrication System and Marine Specific Factors
on Diesel Engine Emissions," Masters Thesis, Departments of Ocean and
Mechanical Engineering, Massachusetts Institute of Technology, 1994.
3. Yoshida, H., et. al. "Diesel Engine Oil Consumption Depending on Piston Ring
Motion and Design" SAE Paper 930995.
4. Lusted, M. R., "Direct Observation of Oil Consumption Mechanisms In a

Production Spark Ignition Engine Using Fluorenscence Techniques," Masters
Thesis, Departments of Ocean and Mechanical Engineering, Massachusetts
Institute of Technology, 1994.
5. Hartman, R. M., "Tritium Method Oil Consumption and Its Relation To Oil Film
Thicknesses In a Production Diesel Engine," Masters Thesis, Departments of
Ocean and Mechanical Engineering, Massachusetts Institute of Technology, 1990.
6. Flaherty, Richard, Private Phone Conversation, CumminsEngine Company, 1995.
7. Hill, S. et. al., "A Systems Approach to Oil Consumption" SAE Paper 910743.
8. Ariga, S. et. al., "On-Line Oil Consumption measurement and Characterization of

an Automotive GasolineEngine by SO2 Method," SAE Paper 920652.
9. Butler, J. W., et al., "Real-Time Measurement of Engine Oil Economy," SAE

Paper 871913, 1987.
10. Yoshifumi, H., et al., "Nissan Oil Econometer Permits the Measurement of Engine
Oil Consumption," SAE Paper 810754, 1981.
11. Lizumi, S., et al., "Measurement of Oil Consumption of Diesel Engine by S-Trace
Method," SAE Paper 860545, 1986.
12. Colvin, A. D., et al., "An Advanced Instrument for the Real Time Measurement of
Engine Oil Economy," SAE Paper 920655, 1992.
13. Schofield, D. M., "Installation and Operation Of a Sulfur Dioxide Tracing Real-
Time Oil Consumption Apparatus," Massachusetts Institute of Technology, May,
1995.
49
14. Ford, E. J., "The Effects of Lubrication System Parameters and Exhaust Aqueous
Injection on Diesel Engine Oil Consumption and Emissions," Masters Thesis,
Departments of Ocean and Mechanical Engineering, Massachusetts Institute of
Technology, 1995.
15. Heywood, J. B., Internal Combustion Engine Fundamentals, McGraw-Hill Book

Company, New York, 1988.
16. Hiruma, M., et al., "Effect of Piston Ring Movement Upon Oil Consumption,"
JSAE Review, March 1983.
17. Schneider, E. W., et al., "Method for Measurement of Piston Ring Rotation in an
Operating Engine," SAE Paper 900224, 1990.
50
Appendix A TESTING RECORDING SHEETS
Data Acquisition Log Sheet
Date Furnace Temp(c)

Temp(C) Short Line Temp(c)
Pressure(mm Hg) 17' Line Temp(c)
Rel Humidity #1 Heat Setting
Torque Offset (N-m) #2 Heat Setting
Span Gas %S #3 Heat Setting
Span Setting on Paper Begin Zero 02 Press(psi)
Span Setting on Paper End Zero Air(psi)
Zero Bypass (Ipm) Air(psi)
Zero Knob Setting NO(psi)
Span Knob Setting S02(psi)
Span Gas Meter RD Furnace(#)
Computer Zero Volt Dryer(#)
Computer Span Volt NO(#)
Computer Zero End 03(#)
Computer Span End
Cell Air Chart
Cell Air Comp
Figure A-I Calibration Data
51
TEST # TEST # TEST # TEST # TEST #
Time Start
Time End
RPM
Load(N-m)(real)
Fuel(cc/min)
Air(g/s)
Head Oil #1(c)
Fuel In #2(c)
Exhaust #5(c)
Air In #6(c)
Exhaust Tnk #7(c)
Oil Out #8(c)
Oil In #10(c)
Coolant Tnk #11(c)
Dyno #12(c)
Coolant In #13(c) _
Cyl Coolant Out #14(c)
Head Coolant Out #15(c)
Cell #16(c)
Bypass Flow (1pm)
Furnace Temp(c)
Figure A-2 Testing Data Sheet
52
Appendix B MATRIX A RESULTS
TEST CO C02 TOTAL AVERAGE AVERAGE AVERAGE AVERAGE AVERAGE

EXH WATER LOAD FUEL AIR OC SOC
(% % molar) (% by weight) (N-m cc/min) (m3/min) (g/hr) (g/kw-hr)
Al 0.070 4.800 4.133 5.328 7.312 0.267 0.148 0.222
A2 0.069 4.800 4.034 5.097 7.340 0.266 0.132 0.207
A3 0.074 4.600 4.115 4.800 7.181 0.266 0.127 0.211
A4 0.110 4.800 4.444 4.462 7.379 0.264 0.136 0.243
B1 0.640 11.500 7.144 16.458 15.490 0.259 0.486 0.235
B2 0.640 11.800 8.514 16.655 15.953 0.258 0.346 0.165
B3 0.640 11.600 8.615 16.198 15.775 0.258 0.314 0.154
B4 0.640 11.600 8.731 15.571 15.693 0.255 0.471 0.241
C1 0.062 4.900 4.111 5.266 15.073 0.484 1.762 1.332
C2 0.065 4.800 4.083 5.101 15.080 0.481 0.918 0.716
C3 0.066 4.900 4.367 5.526 15.216 0.478 1.779 1.281
C4 0.085 4.900 4.297 4.491 15.499 0.477 1.420 1.258
El 0.181 5.800 4.696 4.631 24.709 0.655 1.126 0.704
E2 0.147 5.9013 4.770 4.722 24.494 0.653 0.997 0.611
E3 0.179 5.700 4.889 4.284 24.451 0.656 1.181 0.797
E4 0.206 5.900 4.795 4.000 24.477 0.647 1.241 0.898
F1 0.230 12.200 8.502 16.973 46.449 0.630 2.592 0.442
F2 0.230 12.400 8.672 16.973 47.012 0.627 2.617 0.446
F3 0.230 12.300 8.801 16.536 46.482 0.631 2.521 0.441
F4 0.230 12.400 8.865 15.719 46.906 0.620 2.331 0.429
Figure B-1 Matrix A Data
53
ENGINE OC VS SPEED/LOAD
(ring-pack only)
3
2.5
1.5
.)
H
0
0
1200 1200 2400 3300 3300
LOW HIGH LOW LOW HIGH
CONDITION
Figure B-2 Matrix A Steady-State OC Results
54
ENGINE SPECIFIC OC VS SPEED/LOAD
(ring-pack only)
1.4
1.2
5- 0.8
0.6
0
&n
0.4
0.2
0
1200 1200 2400 3300 3300
LOW HIGH LOW LOW HIGH
CONDITION
Figure B-3 Matrix A Steady-State SOC Results
55
Appendix C MATRIX A REAL-TIME OC RESULTS
NORMALIZED OIL CONSUMPTION

CONDITION A:
1200 RPM LOW LOAD (ring-pack only)
1I I'-'-m-mln·--r-----r--r-·--·--r---·-m-
1 n*nnmnnrnnrrm
0.8
0.6
0.4 -
0.2 - -A4
" . . .. _ . .... .1.
0-
.I
~--A3
z
cO
.. A2
Z: -0.2 - ...........
Al
-0.4 -
-0.6 -
-0.8 -
-1 -
0 120 240 360 480 600
TIME (sec)
NOTE: Data normalizedwith respect to averageOC of all datapointsfor this condition

(NOC = (Xi-Xavg)/Xavg;Xavg= 0.14 g/hr)
Figure C-1 Matrix A Condition A: 1200 Rpm Low Load
56
CONDITION B:
1200 RPM HIGH LOAD (ring-pack only)
1 - I__________·______________··____________ ·_____________________________·_________
0.8 -
0.6 -
0.4 -
. ----- B4
0.2 - ;1r· , .,
, - ---.
- B3
0 0- -- B2
-0.2 - ............ B1
-0.4 -
-0.6 -
-0.8 -
-1 - __ _ _T_ __ __· _ _
0 120 240 360 480 600

TIME (sec)
NOTE: Data normalizedwith respect to averageOC of all data pointsfor this condition
(NOC = (Xi-Xavg)/Xavg;Xavg = 0.40 g/hr)
Figure C-2 Matrix A Condition B: 1200 Rpm High Load
57
CONDITION C:
~.....
I __1_1_1__1__________1 _____I ___ill ___ 111
1
0.8
0.6-
0.4
0.2 - ,· . . .. ...
---C4
Q - C3
0- - C2
-0.2 - ............ C 1
-0.4 -
-0.6 -
-0.8 -
-1 - I __ __ __ I _ _· _ Ir __
2
0 120 240 360 480 6()O
TIME (sec)
NOTE: Data normalizedwith respectto averageOC of all datapointsfor this condition

(NOC = (Xi-Xvg)/Xa,g; Xag = 1.47 g/hr)
Figure C-3 Matrix A Condition C: 2400 Rpm Low Load
58
CONDITION E:
I
I
0.8 -
0.6 -
0.4 -
EA
0.2 -
-- E3
0 0- .... 5· · · ·I··· ·-· ·· · ·· ··· --.··-····· - ·· ··-· _··-/ ,· ··-·-· -· ·-·
--- E2
-0.2 - El
-0.4 -
-0.6 -
-0.8 -
-1 - I ______I
______
____L_ _ _
0 120 240 360 480 600

TIME (sec)
NOTE: Data nomalized with respect to averageOC of all datapointsfor this condition
(NOC = (Xi-Xag)/Xvg; Xg = 1.14 g/hr)
Figure C-4 Matrix A Condition E: 3300 Rpm Low Load
59
CONDITION F:
1
0.8
0.6
0.4
0.2
0
Z
-0.2
-0.4
-0.6
-0.8
-1
0 120 240 360 480 600
TIME (sec)
NOTE: Data normalizedwith respect to averageOC of all datapointsfor this condition

(NOC = (Xi-Xvg)/Xvg; Xag = 2.52 g/hr)
Figure C-5 Matrix A Condition F: 3300 Rpm High Load
60
STRIP CHART RECORDING OF REAL-TIME SO2 ANALYZER OUTPUT
(PROPORTIONAL TO OIL CONSUMPTION)
";-_ i ~~~~ '_

·
~ .~
~ '-
,
-4-- ---
. .. ;
----
. - :-
--------- -, ~ ~ ~ ~~-4
.
--.-
.
;
-~~~~~~~~~~~J
-L
' ·- { :
-
.
, -- - :
.-
-__
'____._.
: i/
_ ____ L '____ I
I'
c
;........
·
.: ,
~
.~~~~ ~~
..
-
~ .~. ~~ . ~~.I. :---~_
-- ~~ -.!~ ~~~~~~~~~~~~~~~~
. .
, . t.......
... . .
i _ _ .
! I . '
~ C,
; , , ... j . .i, ,. , ;
..... _.
__..............
.......... __ ._./_'_
.~.io.
i-
~ic-
' .,'_ !,-
' ....
I
, 41'
-- ..:!7'.-~---
"i.::
.
,
,.. ~~,..
; --- " :-I
'P,7-~--'
---
-!
,.:
.
-''
I .
: *,~~~~~~~~,
'.~
L
-
~
"_-
-_ ' ,. t
! ../i :'~7':
_i., , -. ~-
' . ,
. i_!-~, .".
'°, I,.- ;" :,.I i UMi--'
~ ' o'' ' I ' .,' -'
T ....,'-; ;.... 7 :-,'-: ....F-':... : E;
q;; -, i' ,
' '' ' ',
I
' '
-
'
._..-- ------------r--
i - -r ' :-=
V-
: 1_ ' ' _ ;
3 0:
_,..
,
.__'
. _.
., ~~
!'.= ,
._·
; .
, ...-~....
~ .. ~._'.
, I ,
... ; -i.. ~
. .
~I
I
i
~ I~~
,
.
I I I i I ._. * . j!
0 t0 ao 30 fio 50
OPERATING TIMI (minutes)
Engine rpm & load stabilized

(Start of Test C3)
Figure C-6 Chart Recording of Condition C3
61
STRIP CHART RECORDING OF REAL-TIME SO2 ANALYZER OUTPUT
(PROPORTIONAL TO OIL CONSUMPTION)
-V- 1-.--
! - -I! -I , -I. .
':!.
- '' .-- I
---
. --- - . . C . _ ;--..i t ,, -- . ..
'- - ; - ' L_ I, _., . _ _ . I, ._ _ __ - I .- . --
. I .I
. -7 - -- - _
_ - ; .. .
j _-. .... i - "..
. .i 'I
_
.. . j . , .7 , :--. ... ; I
. .- . .: -I . I 1. ..
--_ _ _ - . .
L.; i I_. I _ _
I :. t /- <,b- I
·· ·- -·-
_ _ , . _ _ ___ __ * _,
i
'' 'i - . :
. ; : .1. . I- '.
, I- 1 - _.
-
. Ii
- .
-:
1~..-',.::.:1:>
J :- , . ; . _, 1:. i i : i
:TI. 7 1.?' · '

L.(
.
~~ ~ ~ __ _______.' __ _ . -1_.
. i
:. '- i .-...-
="::,--
i- -o '
-----
3.
1;--·
-
7 , j
-
C----
; . -'--7-- -- ' ""------
] ° !.
'- ~ ,:~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
, ,' I * ,.' -'___ , ,-' i. .
· I o I
-.-
:.
-
-
I I
, :.i; . , ,- '. : " ,.

- * . ' - ..
':~" :'':
· I 1 ·~
· ·..,..
!::1-..: ', ,.'"! 'i '_
ii. -'.-'
i -: -:,.- : ;'1, ' ",...
I j,,
i __
Ij
_- 1 " ' ,:' '
.. .... j, 7,. , j, -. ~ -- ,, . , -oi , · t - - o? i ! -
.- L C "' I ' "' i' ' '' - '" ''- ' ' 1 1' i i'l ' i 1
...........--- .----- r |-1
f...- .-
- . -,- . -j r
-
_I. , ..--
. ;.i-_. %......·
:......... _. _.-
....
__ _:.',c,,*,,<, ...-,-,..--. --. - ~.......
...- _---4_,: ......... -~- ......... ' '' _'....~'
. ',---
..F
...............
.I..
..............
_,~~~~~
. c
_.'z_ . i., ._
' .. _
:',.. . _ . _ _---i --q
' I_
_
- ' -
_'
. :''I it ? I
_
'~
_.
*
~ i' : '
t-,-'.-.-.......I:
_-'
"','i.---
-_ ..
'
,i
-
·--
..
L .
tc
t
;*"!'i- : -
"'" i ',Fh-~
. _. c. --
------ .............
U.'-
i
. I,,J.~ ·
. I·~L--
-c'- ,t-;,' .-
, · ·~~~~~
. .· ·--· . '-:
'---. . . -- -- i
I I I i I
0 10 AO 36 ¢0
.6PERATING TIME (minutes)
Engine rpm & load stabilized

(Start of Test C2)
Figure C-7 Chart Recording of Condition C2
62
Appendix D MATRIX B RESULTS

(% molar) ( % molar) (% by weight) (N-m) cc/min) (m3/min) (g/hr) (g/kw-hr)
I2 0.146 4.300 3.773 3.769 7.474 0.264 0.098 0.206
I3 0.134 4.500 3.983 3.740 7.188 C.2b3 0.097 0.205
14 0.127 4.600 3.941 4.441 7.129 0.264 0.070 0.126
I5 0.162 4.500 4.094 4.190 7.798 0.264 0.103 0.196
J2 0.560 11.400 8.143 16.095 15.574 0.257 0.558 0.276
J3 0.560 11.500 8.346 15.851 15.458 0.256 0.466 0.234
J4 0.560 11.500 7.961 16.053 15.475 0.255 0.427 0.212
J5 0.560 11.300 8.229 15.295 15.531 0.253 0.189 0.098
L2 0.249 12.100 8.307 16.260 46.846 0.639 5.638 1.003
L3 0.227 12.100 8.490 15.307 46.834 0.634 3.127 0.591
L4 0.250 12.300 8.505 15.913 47.030 0.636 3.422 0.622
L5 0.224 12.400 8.908 15.006 46.961 0.629 3.401 0.656
T2 0.560 11.400 8.190 15.012 15.197 0.254 0.387 0.205
T3 0.560 11.400 8.142 14.719 15.293 0.253 0.237 0.128
U1 0.200 12.100 8.286 16.355 47.180 0.641 4.861 0.860
U2 0.200 12.100 8.431 15.314 46.975 0.630 3.261 0.616
U3 0.200 12.100 8.364 15.008 46.808 0.630 2.890 0.557
Figure D-1 Matrix B Data
63
(ring-pack only)
6
3
6
0
2
0
1200 LOW 1200 HIGH 3300 HIGH
CONDITION
Figure D-2 Matrix B Steady-State OC Results Conditions I, J, L
64
(ring-pack only)
4.5
4
3.5
3
2.5
0
0d
2
1.5
0.5
0
1200 HIGH 3300 HIGH
CONDITION
Figure D-3 Matrix B Steady-State OC Results Conditions T, U
65
(ring-pack only)
1.2
0.8
S.
0.6
Q
0
Co_
0.4
0.2
0
1200 LOW 1200 HIGH 3300 HIGH
CONDITION
Figure D-4 Matrix B Steady-State SOC Results Conditions I, J, L
66
ENGINE SPECIFIC OC VS SPEED/LOAD (ring-
pack only)
0.9
0.8
0.7
0.6
0.5
1o
I-
C) 0.4
cmr
0.3
0.2
0.1
0
1200 HIGH 3300 HIGH
CONDITION
Figure D-5 Matrix B Steady-State SOC Results Conditions T, U
67
Appendix E MATRIX B REAL-TIME OC RESULTS

CONDITION I:
1
0.8
0.6
0.4
0.2
0
-0.2
-0.4
-0.6
-0.8
-1
0 120 240 360 480 600
TIME (sec)
NOTE: Data normalizedwith respect to averagedOC of all datapointsfor this condition

(NOC = (Xi-Xg)/Xavg; Xg = 0.09 g/hr)
Figure E-1 Matrix B Condition I: 1200 Rpm Low Load
68
CONDITION J:
1
0.8
0.6
0.4
0.2 -J5
--- J4
0
0 --- J2
-0.2
-0.4
-0.6
-0.8
-1
0 120 240 360 480 600
TIME (sec)
NOTE: Data normalizedwith respectto averagedOC of all data pointsfor this condition
(NOC = (Xi-Xag)/X,g; Xa,g = 0.41 ghr)
Figure E-2 Matrix B Condition J: 1200 Rpm High Load
69
CONDITION L:
1
0.8
0.6
0.4
- L5
0.2
- A
o0
z - L3
-0.2 ... -.- L2
-0.4
-0.6
-0.8
-1
0 120 240 360 480 600
TIME (sec)
NOTE: Data normalizedwith respectto averagedOCof all data pointsfor this condition
(NOC = (Xi-Xvg)/Xag; Xg = 3.90 g/hr)
Figure E-3 Matrix B Condition L: 3300 Rpm High Load
70
CONDITION T:
1 + -- ~~~~_
0.8
0.6
0.4
Q
0.2 AAJ - A
T2
0 0 - T3
-0.2 - A ~ r In vrwar*
.
...
-0.4
-0.6
-0.8 -I
I - U_ - _
-1
0 600 1200 1800 2400 3000
TIME (sec)
NOTE: Data normalizedwith respect to averagedOC of all data pointsfor this condition
(NOC = (Xi-Xavg)/X,g,; Xag = 0.31 g/hr)
Figure E-4 Matrix B Condition T: 1200 Rpm High Load
71
CONDITION U:
1
0.8
0.6
0.4
0.2 - U1
Uf
0z 0o - U2
-0.2 -U3
-0.4
-0.6
-0.8
-1
0 600 1200 1800 2400 3000
TIME (sec)
NOTE: Data normalizedwith respect to averagedOC of all data pointsfor this condition
(NOC = (Xi-Xvg)/Xavg; Xg = 3.67 g/hr)
Figure E-5 Matrix B Condition U: 3300 Rpm High Load
72
Appendix F MATRIX C RESULTS

(%molar) (%molar) (% by weight) (N-m) ccin) (ccm3/min) (g/hr) (g/kw-hr)
REGAl 0.069 4.800 3.937 4.063 7.795 0.262 0.112 0.219
REGA2 0.069 4.800 3.936 3.253 6.995 0.260 0.198 0.484
REGA3 0.069 4.800 2.949 3.081 7.535 0.258 0.161 0.415
REGA4 0.069 4.800 2.947 3.390 7.167 0.257 0.164 0.385
REGB1 0.560 11.800 8.338 15.770 15.642 0.250 0.376 0.190
REGB2 0.560 11.800 8.336 15.883 15.541 0.252 0.244 0.122
REGB3 0.560 11.800 7.368 16.240 15.366 0.250 0.244 0.120
REGB4 0.560 11.800 8.239 15.622 15.331 0.250 0.279 0.142
REGF1 0.200 12.400 8.564 16.436 46.712 0.627 4.572 0.805
REGF2 0.200 12.400 8.566 16.091 46.530 0.619 4.164 0.749
REGF3 0.200 12.400 7.597 16.921 46.359 0.633 4.492 0.768
REGF4 0.200 12.400 8.663 15.205 46.378 0.612 3.953 0.752
Figure F-1 Matrix C Data
73
4.5
4
3.5
f- 3
2.5
v
L 2
1.5
0.5
0
1200 LOW 1200 HIGH 3300 HIGH
CONDITION
Figure F-2 Matrix C Steady-State OC Results
74
0.9
0.8
0.7
0.6
0.5
v
0.4
0
rr
0.3
0.2
0.1
0
1200 LOW 1200 HIGH 3300 HIGH
CONDITION
Figure F-3 Matrix C Steady-State SOC Results
75
Appendix G MATRIX C REAL-TIME OC RESULTS

CONDITION REGA:
1200 RPM LOW LOAD
1
0.8
0.6
0.4
0.2
0r: 0
-0.2
-0.4
-0.6
-0.8
-1
0 600 1200 1800 2400 3000
TIME (sec)
(NOC = (Xi-Xag)/Xg; Xg = 0.16 g/hr)
Figure G-1 Matrix C Condition REGA: 1200 Rpm Low Load
76
CONDITION REGB:
1200 RPM HIGH LOAD
1
0.8
0.6
0.4
0.2
Q
0
4
-0.2
-0.4
-0.6
-0.8
-1
0 600 1200 1800 2400 3000
TIME (sec)
NOTE: Data normalizedwith respectto averageOC of all data pointsfor this condition
(NOC = (Xi-Xg)/Xag; Xg = 0.29 g/hr)
Figure G-2 Matrix C Condition REGB: 1200 Rpm High Load
77
CONDITION REGF:
3300 RPM HGH LOAD
1
0.8
0.6
0.4
0.2
U
0 0
-0.2
-0.4
-0.6
-0.8
-1
0 600 1200 1800 2400 3000
TIME (sec)
(NOC = (Xi-X.vy)/Xavg;Xg = 4.30 g/hr)
Figure G-3 Matrix C Condition REGF: 3300 Rpm High Load
78
Appendix H COMPARISON OF MATRICES
1.2 - - - ^-r -. ·-·----------- ·-------------------------
I
0.8 l-
[
I-
I3
0.6
-
U
To
Cn
0.4 -
0.2 -
O- ........
1200 LOW 3300 LOW 3300 HIGH
CONDITION
Figure H-1 Total Engine SOC to Ring-Pack Comparison
79
(ring-pack only)
0.8 _I __ _ _____ I ___
___ 1________1___
___
0.7
0 STANDARD OIL
CONTROL RING
0.6 O LESS TENSIONED
OIL CONTROL RING
_' d
0.5
0.4
Q
U0
Cn
0.3 -
-
0.2 -
0.1 -
i
0- R g !
1200 LOW 1200 HIGH 3300 HIGH
CONDITION
Figure H-2 Oil Control Ring Tension Comparison
80
0.8 ~ ---
I-II··I·IV U UWUY YU UU · UYLYY -------··- Y------- Y--------- ------------ -_·_
:·,·
0.7 * HIGH SULFUR 30W OIL i·
::
··.
:·
:· ·'· ··:
o SHELL 10W30 OIL 'REGULAR i·
0.6- OIL' :.. ;.·
·
:;·- :·;.
·
·
0.5 . . ;:
.· .·:
:·i.
·
·:·;· :· i·::;·i·:·.
': ·
:·
0.4 -··· -- .:· .

."
U :.·· ;- : '··-
·- ""- :·. :
cO
3~
c.ar
0.3 - --- :·""
·:·
-- : ·. .:-..' :.
·:·
; · :.
····· :-·;·-·:
· .. : · .·....·
t. ·- :·:·.·:·:··i·'·:·::i-.·
· :
0.2 - :· I.
·s.
: :·
:i
·.n :::·
· :I·:··
0.1 ..
: · .-;
.. A{ .
6·: ::·.:
:.· :·.
.: · ·
0-
./··
:·
:;.
.
·:
_
I . I ·! ·r
1200 LOW 3300 HIGH
CONDITION
Figure H-3 High Sulfur and Commercially Available Oil Comparison
81
82
THESIS PROCESSING SLIP
FIXED FIELD: ill. name
index biblio
COPIES: Acve Aero Dewey (E Hum
Lindgren Music Rotch Science
TITLE VARIES' Or-
NAME VARIES: PI /ac La r
IMPRINT: (COPYRIGHT)
COLII ATION-
""Y
82 p I
*ADD-
DEGREE:M' S DEPT:A ' '
SUPERVISORS:
---
NOTES:
cat'r: date:
page:
O/'
wDEPT:
wYEAR: i f5 DEGREE: A S
*NAME:SCH2E I L Do lm -M

Diesel Engine Instantaneous Oil Consumption Measurements Using The Sulfur Dioxide Tracer Technique

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Diesel Engine Instantaneous Oil Consumption Measurements Using The Sulfur Dioxide Tracer Technique

Hochgeladen von

Copyright:

Verfügbare Formate

Diesel Engine Instantaneous Oil Consumption Measurements Using the Sulfur

Dioxide Tracer Technique

Submitted to the Departments of Ocean Engineering and Mechanical Engineering in

MASTER OF SCIENCE IN NAVAL ARCHITECTURE AND MARINE

© 1995 Douglas M. Schofield. All rights reserved.

Signature of Author __Y__ I_ __ · )__I___

Thesis Supervisor: Victor W. Wong

This work has been supported by the Consortium Of Lubrication in Internal

Chapter 2 EXPERIMENTAL APPARATUS 19

Figure 2-1 Exhaust System 21

Figure A-2 Testing Data Sheet 52

Figure B-2 Matrix A Steady-State OC Results 54

Figure C-7 Chart Recording Of Condition C2 62

Figure E-2 Matrix B Condition J: 1200 Rpm High Load 69

Figure F-2 Matrix C Steady-State OC Results 74

Figure H-2 Oil Control Ping Tension Comparison 80

Figure H-3 High Sulfur and Commercially Available Oil Comparison 81

Table 2-1 Engine Characteristics 19

Table 2-2 Engine Thermocouples 20

Table 3-5 Piston Ring-Pack Characteristics 30

Table 3-7 Test Matrix B 31

Table 3-8 Modified Oil Control Ring Characteristics 32

Table 3-10 Commercially Available Oil Characteristics 33

Table 5-1 Summary of Matrix A Steady-State SOC Results 38

Table 5-2 Summary of Matrix B Steady-State SOC Results 40

Table 5-3 Summary of Matrix C Steady-State SOC Results 43

Table 5-4 Valve Oil Contribution 44

Table 5-6 Matrix C to High Sulfur Oil SCC Comparison 45

SYMBOL DEFINITION UNITS

a # of carbon atoms per fuel molecule

H2 0air Water in ambient air % by wt.

H2Ocombmol Water in exhaust due to combustion % molar

MWexh Molecular weight of exhaust gases g/g-mole

Pa Density of air grams/m 3

Pf Density of fuel grams/cc

1.2 Previous Work

1.3 Oil Consumption Sources

Several different OC measurement techniques have been developed in the past 25

MANUFACTURER: G. CUSSONS LTD.

Table 2-1 Engine Characteristics

Table 2-2 Engine Thermocouples

Figure 2-1 Exhaust System

Figure 2-2 Valve Oil System

2.2 Oil Consumption Apparatus

1/4 " S/S

[M_ THE] FURNACE

Figure 2-3 OC Sampling System

3.1 General Overview Of The Testing Procedure

LATRIX OC OIL TYPE OIL OIL BREAK-

Table 3-1 Overview of Testing

Table 3-2 below shows the details of Matrix A.

NAME SPEED LOAD (bmep) # OF TESTS

Table 3-2 Test Matrix A

3.23 Matrix A Engine Specifics, Controls, and Variables

Table 3-3 Diesel Fuel Characteristics

LUBRICATION BRAND TYPE SULFUR CONTENT

Table 3-4 Oil Characteristics

The ring-pack configuration was the standard as per the manufacturer's

Table 3-5 Piston Ring-Pack Characteristics

The importance of controls cannot be overlooked. The most important control

SPEED ENGINE BMEP FUEL FLOW INJECTION EXHAUST

Table 3-6 Matrix A Variable Specifics

'Fable 3-7 below shows the details of Matrix B.

Signature of Author Y I_ · )I___